← 返回刚才阅读的位置

知识卡片

结构图卡 · 8825.d · 生成模型:GPT-6

从输入到解释的层次关系

理解等级:中级

视觉系统从有限的光信号形成任务相关解释,需要区分物理输入、表示、算法和实现。本图概括Marr历史表示路线,并不声称所有现代模型具有相同中间层。

结构图:

flowchart LR
 A[图像强度] --> B[原始草图 边缘与组织]
 B --> C[二维半 深度与表面方向]
 C --> D[三维形状与空间关系]
这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      In David Marr's approach, vision was implemented as a sequence of representations, starting from the image represented as a sequence of pixel intensities, and then transforming each representation into another one that extracts more explicit information about the geometry and objects present in the world. The sequence of representations proposed by Marr was the following:

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第01章《视觉的挑战》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c001.xhtml) - 结论依据:摘录中关于从输入到解释的层次关系的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:In David Marr's approach, vision was implemented as a sequence of representations, starting from the image represented as a sequence of pixel intensities, and then transforming each representation into another one that extracts more explicit information about the geometry and objects present in the world. The sequence of representations proposed by Marr was the following: